|
Subject:Settings for good recording and managing wavs
Posted by: mrjohn
Date:2/19/2004 12:15:48 AM
Hi! I try to record some old cassette tapes and vinyls on computer to burn them later on cd. But I have some trouble to set right preferences for good recordings. Can you tell me what settings and actions should I use/take to get best results for later to record wavs on cd. I use hifi output connected to sound card, and phono with riaa-preamp (connected via hifi and sometimes directly to sound card; what is the best way?). 1. What level should I use for recording. In which interval should be peaks and VU/PPM levels to avoid clipping and that I can later get the best sound. 2. To which level should I set normalize (is that 20db or other), or is better to use change volume. I figure out, that I must use normalize separate on left and right channel (some old tapes have different volume on left and right channel). Can I use normalize first on left, right and then on both channels? 3. Which functions after recording should I use and in which order. Is that dc-offset, normalize, fade-in/out. For vinyl restoration is restoration better before or after normalize or volume change. 4. Which settings are required to clear noise and not to change quality of sound. 5. Some old tapes have very little high frequences. Is there a wav to strength them (what left of them). 5. I use Nero burn to burn wav files on audio-cd. Should I use features (normalize, fade,..) in nero or is better to use functions in Sound Forge. Thanx, Robert |
|
Subject:RE: Settings for good recording and managing wavs
Reply by: Chienworks
Date:2/19/2004 5:17:54 AM
Just a few answers ... 1. The ideal is to come as close to 0dB as possible without hitting it. However, since digital recording has a much greater dynamic range than the signal you're getting through your sound card, you probably have a lot of leeway. I usually try to get my peaks somewhere above -18dB while recording. 2. You should normalize your peaks to just below 0dB. I have normalize set to -0.1dB. Rip a commercial pop CD and you'll see that most of them are hugging 0dB most of the time. Normalizing on each channel separately isn't a bad idea if there is a big difference in level. It's not necessary to renormalize both together after you've done each separately though. 3. DC offset usually isn't necessary. I wouldn't use it unless it's off by more than a few hundred. Generally normalizing should always be the last step. If you've recorded your signal strong enough (peaks over -18dB) then you shouldn't have to do any normalization or volume changes before restoration processes. 4. That depends. Experiment to find out. It will probably be different for each recording, and may even be different for different parts of the same recording. 5. EQ will help this. Just remember that you can't get blood from a stone. Raising high frequencies will also tend to raise hiss type noises as well. You may want to do this step before using noise reduction. 5 (again?). SoundForge will probably do a better job on these tasks than Nero will. Above all else, do as little processing as possible! Every single thing you do to the recording will degrade the sound. The fewer processes you do the better the result will be. This even includes volume adjustments. Normalizing once at the end is the only time you should be changing the volume; adjusting the volume before or after any other steps is unnecessary and only degrades the signal more. |
|
Subject:RE: Settings for good recording and managing wavs
Reply by: FrenchTouch
Date:2/20/2004 12:20:58 AM
I agree with you Chienwork. mr john, I suggest you to use Noise Reduction. The result is often better if you do 2 little Noise Reduction in a plug-in chainer than 1 important. It's the same thing with "Vinyl Restoration", that you'll can use to remove clics & crackles. Finally, normalize your sound to -0.2 Db. Too much normalizations will degrade your sound. Bye-Bye ! |
|
Subject:RE:Normalize peaks or rms
Reply by: mrjohn
Date:2/20/2004 2:33:15 PM
Hi! When I normalize sounds to peak 0, then copy them to cd: songs are not equal volume, but when I normalize to rms -20, songs on cd were all equal volume (but peaks are different). In your answers, you tell me that I should use normalize peaks. Now I dont't know, what normalize should I use to get right levels for cd. Can you tell me, what actions should I take. I wrote some examples, from my recordings. Thanx, Robert status ---after normalize---- Recorded peak to 0.0 rms to -20 ----- -------- ----------- ---------- peak -7.9 0.0 -6.5 ----- -------- ----------- ---------- rms -21.3 -13.4 -20.0 ----- -------- ----------- ---------- ----- -------- ----------- ---------- peak -9.2 0.0 -6.7 ----- -------- ----------- ---------- rms -22.5 -13.2 -20.0 ----- -------- ----------- ---------- ----- -------- ----------- ---------- peak -5.2 0.0 -5.3 ----- -------- ----------- ---------- rms -19.8 -14.6 -20 ----- -------- ----------- ---------- ----- -------- ----------- ---------- peak -1.6 0.0 -8 ----- -------- ----------- ---------- rms -13.6 -11.9 -20 ----- -------- ----------- ---------- |
|
Subject:RE: RE:Normalize peaks or rms
Reply by: BobMarvin
Date:2/20/2004 5:22:53 PM
I'm also using SF and Nero 6. I've gotten excellent results transfering all sorts of material by peaking at -4 DB in SF (I work in the broadcast industry and our metering is +4 based. When I play a commercial CD it peaks 0 on our p+4 meters) and by peaking -4 DB in SF and setting the normalizing in Nero at 10 (20 is default) I get a good quality CD every time. Do not save the metadata. If you do you won't be able to burn DAO w/out problems. The fun part of Nero is that you can save data for each file, like MP3's when you're burning a wave or cda file and Nero allows for different periods of silence between every cut. For instance some cuts can seg while others can have a 2 (or whatever time you chose) second pause between cuts. It's a great compliment to SF. |